Test helm all k8s available versions

.github/actions/run-integration-tests/action.yaml

@@ -5,12 +5,15 @@ runs:
   steps:
     - name: Run Integration tests
       run: |
+        kubectl wait --for=condition=ready --timeout=60s pod -l app=timeseries-mock-service -n test-namespace
+
         kubectl create job --from=cronjob/integration-test integration-test-manual -n test-namespace
 
         kubectl wait --for=condition=ready --timeout=60s pod -l job-name=integration-test-manual -n test-namespace
         kubetail -l job-name=integration-test-manual -n test-namespace &  
 
         # Wait for either complete or failed
+        echo "Waiting for tests to complete"
         while true; do
             if kubectl wait --for=condition=complete --timeout=0 jobs/integration-test-manual -n test-namespace 2>/dev/null; then
                 job_result=0
@@ -21,8 +24,7 @@ runs:
                 job_result=1
                 break
             fi
-
-            echo "Waiting for tests to complete"
+            echo -n "."
             sleep 3
         done
 
@@ -38,7 +40,7 @@ runs:
                 kubectl logs -n test-namespace $pod -c $container > pod-logs/$container.txt
               done
             done
-
+            mv raw_bodies_dump.txt pod-logs/
             exit 1
         fi
 

.github/workflows/buildAndDeploy.yml

@@ -31,6 +31,18 @@ jobs:
       - name: Deploy skaffold
         uses: ./.github/actions/deploy-skaffold
 
+      - name: Create k8s Kind Cluster
+        uses: helm/[email protected]
+        with:
+          version: v0.24.0
+          kubectl_version: v1.31.0
+          node_image: kindest/node:v1.31.0
+          cluster_name: kind
+
+      - name: Add dependency chart repos
+        run: |
+          helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
+
       - name: Generate docker image tag
         id: generate-tag
         run: echo "::set-output name=value::v${{ github.run_number }}-$(git rev-parse --short HEAD)"
@@ -41,15 +53,24 @@ jobs:
       - name: Test
         run: skaffold test --build-artifacts=/tmp/tags.json
 
-      - name: Deploy kubernetes
-        uses: ./.github/actions/deploy-kubernetes
+      - name: Copy image to Kind cluster
+        run: |
+          kind load docker-image $(jq -r '.builds[] | select(.imageName=="integration-test") | .tag' /tmp/tags.json)
+          kind load docker-image $(jq -r '.builds[] | select(.imageName=="swi-k8s-opentelemetry-collector") | .tag' /tmp/tags.json)
+          kind load docker-image $(jq -r '.builds[] | select(.imageName=="swi-k8s-opentelemetry-collector-tests") | .tag' /tmp/tags.json)
 
-      - name: Add dependency chart repos
+      - name: Setup kubetail
         run: |
-          helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
+          curl -Lo kubetail https://raw.githubusercontent.com/johanhaleby/kubetail/master/kubetail && sudo install kubetail /usr/local/bin/
+        shell: bash
 
       - name: Deploy services using Skaffold
-        run: skaffold deploy --build-artifacts=/tmp/tags.json
+        run: skaffold deploy --build-artifacts=/tmp/tags.json --kube-context kind-kind
+
+      - name: Show all resources
+        if: ${{ failure() }}
+        run: |
+          kubectl get all -n test-namespace
 
       - name: Run integration tests
         uses: ./.github/actions/run-integration-tests

.github/workflows/buildAndTestHelm.yml

@@ -51,15 +51,32 @@ jobs:
     name: Helm e2e tests
     outputs:
       image_tag: ${{ steps.generate-tag.outputs.value }}
+
+    strategy:
+      matrix:
+        # Kubernetes versions to test on
+        kubernetes_version:
+          - v1.25.16
+          - v1.26.15
+          - v1.27.16
+          - v1.28.13
+          - v1.29.8
+          - v1.30.4
+          - v1.31.0
     steps:
       - name: Checkout
         uses: actions/checkout@v4
 
       - name: Deploy skaffold
         uses: ./.github/actions/deploy-skaffold
 
-      - name: Deploy kubernetes
-        uses: ./.github/actions/deploy-kubernetes
+      - name: Create k8s Kind Cluster
+        uses: helm/[email protected]
+        with:
+          version: v0.24.0
+          kubectl_version: ${{ matrix.kubernetes_version }}
+          node_image: kindest/node:${{ matrix.kubernetes_version }}
+          cluster_name: kind
 
       - name: Add dependency chart repos
         run: |
@@ -68,12 +85,29 @@ jobs:
       - name: Build
         run: skaffold build -p=ci-helm-e2e --file-output=/tmp/tags.json
 
+      - name: Copy image to Kind cluster
+        run: |
+          kind load docker-image $(jq -r '.builds[] | select(.imageName=="integration-test") | .tag' /tmp/tags.json)
+
+      - name: Setup kubetail
+        run: |
+            curl -Lo kubetail https://raw.githubusercontent.com/johanhaleby/kubetail/master/kubetail && sudo install kubetail /usr/local/bin/
+        shell: bash
+
       - name: Deploy services using Skaffold
-        run: skaffold deploy -p=ci-helm-e2e --build-artifacts=/tmp/tags.json
+        run: |
+          skaffold deploy -p=ci-helm-e2e \
+            --build-artifacts=/tmp/tags.json  \
+            --kube-context kind-kind
 
+      - name: Show all resources
+        run: |
+          kubectl get all -n test-namespace
+
       - name: Run integration tests
         uses: ./.github/actions/run-integration-tests
 
+
   helm_test_auto_update_against_last_published:
     runs-on: ubuntu-latest
     name: Helm auto update test

tests/integration/expected_metric_names.txt

@@ -138,7 +138,6 @@ k8s.node.network.transmit_packets_dropped
 k8s.node.pods
 k8s.node.status.condition.diskpressure
 k8s.node.status.condition.memorypressure
-k8s.node.status.condition.networkunavailable
 k8s.node.status.condition.pidpressure
 k8s.node.status.condition.ready
 k8s.persistentvolumeclaim.status.phase

tests/integration/test_log_collection.py

@@ -1,11 +1,12 @@
 import pytest
 import os
+import json
 from test_utils import get_all_bodies_for_all_sent_content, retry_until_ok, run_shell_command
 
 endpoint = os.getenv("TIMESERIES_MOCK_ENDPOINT", "localhost:8088")
 url = f'http://{endpoint}/logs.json'
 pod_name = 'dummy-logging-pod'
-tested_log = '!!testlog!!'
+tested_log = 'testlog-swo-k8s-collector-integration-test'
 
 def setup_function():
     run_shell_command(f'kubectl run {pod_name} --image bash:alpine3.19 -n default -- -ec "while :; do echo \'{tested_log}\'; sleep 5 ; done"')
@@ -18,12 +19,17 @@ def test_logs_generated():
 
 def assert_test_log_found(content):
     raw_bodies = get_all_bodies_for_all_sent_content(content)
-    test_log_found = any(f'{tested_log}\n' in body for body in raw_bodies)
+    test_log_found = any(f'{tested_log}' in body for body in raw_bodies)
     return test_log_found
 
 def print_failure(content):
     raw_bodies = get_all_bodies_for_all_sent_content(content)
     print(f'Failed to find {tested_log}')
-    print('Sent logs:')
-    print(raw_bodies)
+    print('All logs in raw_bodies_dump.txt')
+    #print(raw_bodies)
+    # Dump the raw_bodies to a file
+    with open('raw_bodies_dump.txt', 'w') as file:
+        # Convert the list to a JSON string for better formatting
+        json.dump(raw_bodies, file, indent=4)
+
 

tests/integration/test_metric_collection.py

@@ -46,7 +46,7 @@ def test_expected_otel_message_content_is_generated(file_name):
     resource_attributes = test_case["resource_attributes"]
     metrics = test_case["metrics"]
 
-    print(f'Checking metrics {metrics} with resource attributes {resource_attributes}')
+    print(f'Checking metrics {[item['name'] for item in metrics]} with resource attributes {resource_attributes}')
 
     retry_until_ok(url, 
                    lambda content: assert_test_contain_expected_datapoints(content, metrics, resource_attributes),
@@ -205,7 +205,7 @@ def assert_test_contain_expected_datapoints(content, metrics, resource_attribute
                         # Loop through each metric
                         for metric in scope["metrics"]:
                             if metric["name"] == metric_in_test_case["name"]:
-                                print(f'Found metric {metric_in_test_case["name"]} in resource group')
+                                print(f'Found metric {metric_in_test_case["name"]}')
                                 test_case_passed = True
 
                                 # Default to empty list if 'attributes' key is not present
@@ -247,7 +247,7 @@ def assert_test_contain_expected_datapoints(content, metrics, resource_attribute
                 break  # Metric found, break the resource loop            
 
         if not test_case_passed:
-            return (False, f'Failed to find metric {metric_in_test_case["name"]} in resource group')
+            return (False, f'Failed to find metric {metric_in_test_case["name"]}')
 
     return (True, '')
 

tests/integration/test_utils.py

@@ -30,7 +30,7 @@ def get_all_resources_for_all_sent_content(content):
     return [get_all_log_resources(log_bulk) for log_bulk in log_bulks]
 
 
-def retry_until_ok(url, func, print_failure, timeout = 240):
+def retry_until_ok(url, func, print_failure, timeout = 600):
     start_time = time.time()
     last_exception = None
     last_error = ''
@@ -48,8 +48,8 @@ def retry_until_ok(url, func, print_failure, timeout = 240):
             print(e, traceback.format_exc())
 
         if response is not None and response.status_code == 200:
-            if( last_error == ''): 
-                print("Successfully downloaded!")
+            if( last_error != ''): 
+                print(last_error)
             result = func(response.content)
             if( type(result) != tuple):
                 is_ok = result